Validation of postmortem PSA as an accurate reflection of the deceased's actual PSA is a necessary step before major epidemiologic studies can be conducted on cadaver prostates. Our group has extensive experience with the conduct of large scale epidemiologic studies on autopsy specimens and have recently described the prevalence of prostate cancer in young white and African American males. We are initiating a study to evaluate the ability of PSA and its related forms such as free PSA (fPSA), complexed PSA (cPSA), hK2, etc. to detect microscopic, occult malignancies. It is important to characterize the ability of PSA to detect these lesions, because they may need to be diagnosed in young males, whereas they may not be biologically important in elderly men. Unfortunately we can only identify from surgical series those lesions which were already detected by PSA, and do not have knowledge of those tumors which were not identified by PSA (false negative results). This information is better derived from autopsy series, provided postmortem PSA is an accurate reflection of the individual's pre-mortem values. The purpose of the current study is to seek validation of this hypothesis. We propose to evaluate total PSA (tPSA), fPSA, cPSA and hK2 levels in 60 men autopsied at our institution and who have serum samples available from prior to death to analyze for PSA. Blood samples taken prior to death will be compared to PSA measurements on samples obtained at various time points up to 48 hour following death.